Citation
Comparisons of intermediate biomarkers during carcinogenesis and intervention with grape seed extract

Material Information

Title:
Comparisons of intermediate biomarkers during carcinogenesis and intervention with grape seed extract
Creator:
Spears, Erick Alan
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
24 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Carcinogenesis ( lcsh )
Grapes -- Therapeutic use ( lcsh )
Biochemical markers ( lcsh )
Biochemical markers ( fast )
Carcinogenesis ( fast )
Grapes -- Therapeutic use ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 23-24).
Thesis:
Biology
General Note:
Department of Integrative Biology
Statement of Responsibility:
by Erick Alan Spears.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
45545100 ( OCLC )
ocm45545100
Classification:
LD1190.L45 2000m .S63 ( lcc )

Downloads

This item has the following downloads:


Full Text
COMPARISONS OF INTERMEDIATE BIOMARKERS DURING
CARCINOGENESIS AND INTERVENTION WITH GRAPE SEED EXTRACT
by
Erick Alan Spears
B.A., University of Colorado, 1994
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Biology


2000 by Erick Alan Spears
All rights reserved.


This thesis for the Master of Arts
degree by
Erick Alan Spears
has been approved
by
Margaret E.
nausek
7


Spears, Erick Alan (M.A., Biology)
Comparisons of Intermediate Biomarkers During Carcinogenesis and
Intervention with Grape Seed Extract
Thesis directed by Associate Professor Bradley J. Stith
ABSTRACT
The SENCAR mouse skin model of carcinogenesis provides the
necessary framework for studying the chemical and physiological indicators
of carcinogenesis. Intermediate biomarkers are chemical indicators or
physiological responses that mark intermediate outcomes in carcinogenesis.
These biomarkers are highly predictive of cancer development. The
SENCAR mouse model was used to understand trends in intermediate
biomarkers over an extended period of carcinogenesis as well as to assess
the inhibitory effects of a natural source antioxidant on selected intermediate
biomarkers. Experimental groups of thirty mice were treated one time with
10 nmol DMBA followed by twice weekly treatment with 2 pg TPA for four
weeks. Two groups were treated with a commercial source grape seed
extract ten minutes prior to TPA, and one of these groups was treated with
grape seed extract prior to DMBA. All groups were sacrificed five mice at a
time every two weeks beginning just prior to the first TPA treatment. All skin
samples were analyzed for selected intermediate biomarkers, epidermal and
dermal hyperplasia, myeloperoxidase activity, 8-hydroxy-2-deoxyguanosine,
and Ha-ras oncogene. The commercial source grape seed extract did not
modulate intermediate biomarker levels in this study. The treatment regimen
specified for this study was sufficient to induce significant increases in levels
of each of the intermediate biomarkers. While significant increases were
observed, by the end of the study biomarker levels had returned to baseline
levels.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed


Brad.
IV


CONTENTS
Figures..............................................................vii
Chapter
1. Introduction ....................................................1
2. Methods .........................................................5
2.1 SENCAR mouse skin carcinogenesis ................................5
2.2 Histological evaluation of epidermal and dermal hyperplasia......6
2.3 Myeloperoxidase activity assay...................................6
2.4 DNA extraction...................................................7
2.5 8-Hydroxy-2-deoxyguanosine analysis.............................8
2.6 Ha-ras oncogene analysis.........................................9
2.7 Data correlation................................................10
2.8 Reagents........................................................10
3. Results.........................................................11
3.1 Inflammatory and hyperplastic responses.........................11
3.1.1 Epidermal and dermal hyperplasia................................12
3.1.2 Myeloperoxidase activity........................................14
3.2 DNA analysis ...................................................15
3.2.1 8-OH-dG ........................................................15
V


3.2.2 Ha-ras .....................................................16
3.3 Trend comparisons ..........................................18
3.4 Papilloma development.......................................19
4. Discussion..................................................20
References .......................................................23


FIGURES
Figure
3.1 Epidermal and dermal hyperplasia results...................13
3.2 MPO activity in epidermal tissue from SENCAR
mice treated with acetone or different combinations
of DMBA, TPA and GS .......................................15
3.3 8-OH-dG /105 dG ratios for epidermal tissue from
SENCAR mice treated with acetone or different combinations
of DMBA, TPA and GS .......................................16
3.4 Mutant / wild-type ratios for A-> T transversions at
codon 61 of the Fla-ras oncogene in epidermal tissue
from SENCAR mice treated with acetone or different
combinations of DMBA, TPA and GS ..........................17
3.5 Percent change from baseline with time for all intermediate
biomarkers assayed.........................................19


1. Introduction
Cancer is a disease characterized by the uncontrolled proliferation of
otherwise normal cells (Cooper, 1992). There are many different types of
cancer corresponding to the many cell types found in complex animal
systems. Neoplastic transformation is initiated by an insult to the genetic
code, a mutation. Mutations to genes that are directly involved in regulation
of the cell cycle may ultimately result in uncontrolled cell proliferation. Once
the mutated cell divides, the mutation is locked-in to the genome of all
progeny. Thus, cancer cells divide to form more cancer cells, and
eventually invade and damage healthy tissues.
Carcinomas, cancers that arise from epithelial tissues, are of particular
importance in the field of cancer research. Approximately 90% of all human
cancers are carcinomas (Cooper, 1995). Due to the rapid cell proliferation
that occurs in most normal epithelial tissues, carcinomas are also among the
most dangerous cancers.
The SENCAR mouse skin model of carcinogenesis provides the
necessary framework for studying the chemical and physiological indicators
of cancer development. This model of skin carcinogenesis is a stepwise
process in which genetic damage is accumulated at each step (DiGiovanni,
1992). The first step, involving an initial insult to the DNA at a target gene,
is referred to as initiation. The initiating insult alone is not enough to cause
tumorigenesis. A second step, known as promotion, must occur in order for
the skin cells to become cancerous. The promoting agent causes the cells
to undergo clonal expansion and form premalignant lesions called
l


papillomas. The final step, progression, is the accumulation of further
genetic changes causing papillomas to become malignant.
SENCAR mice are a particular strain genetically developed to be
highly sensitive to this type of multistage carcinogenesis using a variety of
chemical carcinogens (Boutwell, 1964; DiGiovanni, 1980). Treatment with
any known carcinogen, at the appropriate dosage and repetition, leads to
inflammation and regenerative hyperplasia, dysplasia, papillomas, and
finally, basal and/or squamous cell carcinomas (Slaga, 1987; DiGiovanni,
1992).
A major obstacle to the study of cancer pathogenesis is the fact that
tumors begin as normal tissue cells. Tumors are not foreign diseases such
as pathogenic bacteria that differ greatly in structural and genetic make-up
from their hosts. Often cancers are morphologically and physiologically very
similar to normal tissue cells until the late stages of disease. As early
detection of cancer has been directly correlated to survival and long-term
prognosis, the search for early signs of carcinogenesis has become a
priority in the field of cancer research. These early signals of
carcinogenesis have been termed biomarkers (tumor markers,
carcinogenesis markers). Biomarkers may be chemical indicators, such as
unusual proteins, or physiological responses, such as local inflammation
and hyperplasia (Slaga, 1987; DiGiovanni, 1992). These indicators are very
useful in cancer research, as tools for early detection of human tumors, and
in the many in vitro and in vivo models used to study carcinogenesis.
Intermediate biomarkers occur prior to and are highly predictive of tumor
development. They are observed in vitro and in vivo as a means of (1)
identifying individuals with high susceptibility to cancer, (2) measuring
compliance with interventions, and (3) to marking intermediate outcomes
and possibly simplifying experiments (Hanausek, 1995).
2


Cancer prevention has come to the forefront of cancer research
recently. Chemoprevention has become a very popular subject in cancer
prevention studies. Chemoprevention is defined as the use of chemical
agents, either natural or synthetic, to reverse, suppress, or prevent the
establishment of a malignant growth (Vogel, 1995). These chemical agents
may act to inhibit damage to DNA by blocking the action of carcinogens or
interfere with the effects of a promoting agent by suppressing the clonal
outgrowth of transformed cells (Cooper, 1992). Antioxidants, chemicals that
inhibit oxidative damage to biological molecules, have been recognized as
important chemopreventive agents. Oxidative damage has been implicated
in both tumor initiation and promotion (Floyd, 1990). Well established in vivo
models, such as the SENCAR mouse skin model, have linked oxidative
damage to tumor development (DiGiovanni, 1992). In further studies,
intermediate biomarker levels and ultimate tumor development have been
inhibited by treatment with well-known antioxidants (unpublished results).
Recently, natural source antioxidants such as phytochemical extracts have
shown promise as anti-carcinogenic agents. Of these phytochemical
extracts, the polyphenols found in grape seed extracts have been shown to
decrease tumor incidence in a two-stage model of carcinogenesis (Zhao,
1999).
In order to evaluate the effects of both initiation and promotion on
several intermediate biomarkers, a multistage model of carcinogenesis was
employed. The initiating agent used was 7,12-dimethylbenz[a]anthracene
(DMBA). DMBA belongs to a class of chemical carcinogens known as
polycyclic aromatic hydrocarbons (PAH). In biological systems, PAHs are
metabolized to diol epoxides which bind to DNA and leave behind oxidative
lesions (Dipple, 1984). Therefore, DMBA is an effective initiator in mouse
skin models of multistage carcinogenesis. A one time dose of 10 nmol
3


DMBA (initiating dose), followed by extensive treatment with a promoting
agent is sufficient to cause carcinoma development in SENCAR mice
(Boutwell, 1964).
The most widely studied class of promoters in mouse skin models is
the phorbol esters. The most potent phorbol ester is 12-0-
tetradecanoylphorbol-13-acetate (TPA; DiGiovanni, 1992). TPA mimics the
action of diacylglycerol in cells and initiates the protein kinase C (PKC)
pathway. Among the numerous effects of PKC activation is the stimulation
of cell growth and proliferation (Morgan, 1989). While a single initiating
dose of DMBA is not sufficient to cause tumor formation, four weeks of twice
weekly treatment with TPA will ultimately result in carcinoma formation
(Slaga, 1989). There is also a safety advantage to this two-stage treatment
model. A single, low dose treatment with carcinogen is administered, and
further treatments employ a compound, TPA, that is intrinsically non-
carcinogenic.
The purpose of this study was two-fold. First, the behaviors of several
intermediate biomarkers were monitored over an extended period of
carcinogenesis. Secondly, the effect of co-treatment with a commercial
source Grape Seed Extract (GS) on the behaviors of these biomarkers was
evaluated. The use of a two-stage model allows for the observation of the
effects of both initiation and promotion on these intermediate biomarkers.
4


2. Methods
2.1 SENCAR mouse skin carcinogenesis
Seven-week old, pathogen free, female SENCAR mice were
purchased from the National Cancer Institute (NCI, Frederick, MD). Mice
were randomized by weight and separated into four groups of 30 mice.
They were then shaved and topically treated with different combinations of
carcinogen, antioxidant, and solvent. Groups 1 and 2 served as negative
and positive controls, respectively. Group 1 was treated with acetone only
as this was the solvent for all treatment compounds. Groups 2,3 and 4 were
treated with initiator, 10 nmol DMBA, and left untreated for one week.
Following the one week washout period, 2 pg TPA was administered twice
weekly for four weeks. Acetone treatments were administered at the same
times as the DMBA and TPA treatments. Groups 3 and 4 were treated with
GS no less than ten minutes prior to treatment with TPA. Group 3 was also
treated with GS prior to DMBA treatment, while group 4 was not.
Animals were sacrificed at two-week intervals beginning just prior to
the first treatment with TPA and continuing for six weeks following cessation
of treatment. At sacrifice the shaved dorsal skin section was removed. A
one square centimeter section was removed from the center of the skin,
preserved in 10% buffered formalin, and embedded for histological
preparation. Two 11 mm diameter punches were removed directly anterior
and posterior to the histological section and frozen in liquid nitrogen. The
remaining skin was frozen in liquid nitrogen. All frozen sections were stored
at -70C until analysis.
5


2.2 Histological evaluation of epidermal
and dermal hyperplasia
The formalin preserved sections of the dorsal skin were histologically
prepared, stained with hematoxalin and eosin, and examined for epidermal
and dermal hyperplasia. Epidermal and dermal thicknesses were evaluated
using an Olympus BH-2 microscope with a 20.4 mm diameter measuring
eyepiece and a micrometer disc. For each sample, 20 measurements were
made and an average thickness in micrometers (pm) was calculated.
2.3 Myeloperoxidase activity assay
Quantification of myeloperoxidase (MPO) activity was carried out
using the methods described by Bradley et al., 1982, and Trush et al., 1994.
Briefly, MPO was extracted from 11 mm diameter punches of skin. The skin
was minced and homogenized for twenty seconds in potassium phosphate
buffer containing 0.5% hexadecyltrimethylammonium bromide (HTABr), pH
6.0. The homogenized solutions were freeze-thawed and sonicated three
times. After the final sonication, the suspensions were centrifuged at
15.000 g and 4C for twenty minutes. The resulting supernatant was
assayed for MPO activity at 460 nm on a Packard, SpectraCount 96 well
plate reader. The assay was performed in a potassium phosphate buffer
containing 0.5% HTABr, 0.167mg/mL o-dianisidine dihydrochloride,
0.0005% hydrogen peroxide and 10 pL of sample supernatant to a final
volume of 300 pL. MPO activity in milliunits (mil) was calculated by
comparison with curves from known concentrations of MPO standard, 0.3 -
10.0 mU.
6


2.4 DNA extraction
The final two analyses required DNA purification. DNA was purified
by repetitive phenol, chloroform and ether extractions followed by ethanol
precipitation. The largest section of frozen skin was thawed and the
epidermal layer was isolated and removed by trypsinization. The thawed
skin was placed hair side up in a petri dish and scalpel incisions were made
approximately 0.5 to 1.0 cm apart. The skin was completely covered in
0.25% trypsin and incubated at 37C for two hours. The trypsin reaction
was stopped by removal of the trypsin solution and replacement with an
equivalent amount of fetal bovine serum (FBS). The epidermal layer was
lightly scraped off with a sterile scalpel and transferred to a 13 mL, round-
bottom, centrifuge tube. The scraped epidermal tissue was frozen in liquid
nitrogen and stored at -70C until DNA extraction.
The scraped epidermal tissue was gradually thawed and suspended
in 3 mL of lysis buffer consisting of 5% SDS, 10 mM NaCI, 10 mM Tris, 10
mM EDTA, and 100 jig/mL of Proteinase K. The suspension was incubated
at room temperature overnight. The first cycle of extraction was performed
with 1x volumes of phenol, chloroform / isoamyl alcohol (CIA, 24:1) and
ether. Tubes were closed and shaken for ten minutes following the addition
of phenol and five minutes following the addition of CIA and ether. After
shaking, extracts were spun for ten minutes at 10,000 rpm while in phenol
and CIA, and 3,000 rpm while in ether. The organic phases from these
extractions were discarded. DNA was precipitated twice by the addition of a
1/1 Ox volume of 4M sodium acetate, and a 2-3x volume of cold 95%
ethanol. After the second precipitation, the DNA was dissolved in 600 piL of
7


1xTris-EDTA (TE) and stored overnight at 4C. Prior to a second cycle of
extraction, a 1/1 Ox volume of 1 mg/mL RNase A was added and samples
were incubated at 37C for two hours. The second cycle of extraction
involved the substitution of phenol / chloroform / isoamyl alcohol (25:24:1)
for phenol and proceeded as described above. After extraction, the samples
were again precipitated twice with 4M sodium acetate and 100% ethanol
and subsequently dissolved in 500 pL of TE.
Purified DNA was analyzed spectrophotometrically for purity. Optical
density (OD) readings were taken at 260 nm and 280 nm. An OD 260/OD
280 ratio of 1.8 or higher was considered acceptable purity. The
concentration of purified DNA was determined from the OD 260 reading.
The DNA was also analyzed for degradation by 0.7% agarose gel
electrophoresis.
2.5 8-Hydroxy-2-deoxyguanosine analysis
8-Hydroxy-2-deoxyguanosine (8-OH-dG) in treated skin was
quantified using the isocratic high pressure liquid chromatography (HPLC)
method described by McCabe etal., 1997. Approximately 100 pg of DNA
from each sample was hydrolyzed with Nuclease P1 and Alkaline
Phosphotase and injected on column. The chromatographic conditions
were similar to those described previously with minor changes. The mobile
phase was 100 mM sodium acetate in 4% methanol, pH 5.2. All reagents
were HPLC grade and milli-Q water was further purified using a Waters Sep-
Pak solid phase extraction column. Isocratic analysis took place on a
Shimadzu LC-600 pump connected to a Spectrasystem AS3500
autosampler. A YMCbasic S3, 4.6 mm x 150 mm column was used for
separation. The oxidized guanine adduct, 8-OH-dG, was detected using a
8


CoulArray electrochemical detection system (ESA, Inc., Chelmsford, MA).
The cell potentials were set at 50/100/150/230/285/325/400/0. The eighth
cell was adapted to a connection with a Shimadzu SPD-6A UV detector set
at 254 nm. Non-oxidized 2-deoxyguanosine (dG) was quantified by UV
detection. Sample adduct concentrations were calculated from standard
curves of 8-OH-dG, 0.1 1.5 pmol, and dG, 0.5 15.0 nmol. All data were
analyzed using ESA CoulArray for Windows software and expressed in
terms of the ratio 8-OH-dG/105dG.
2.6 Ha-ras oncogene analysis
Mutation of the Ha-ras oncogene in the treated samples was assayed
using a polymerase chain reaction (PCR) protocol derived from Nelson et
al., 1992. The mutant and wild-type genes were amplified with mutation
specific (5-CATGGCACTATACTCTTCTA) and wild-type specific (5-
CATGGCACTATACTCTTCTT) 3-primers. The same 5-primer (5-
CTAAGCCGTGTTGTTTTGCAGGAC) was used in both mutant and wild-
type reactions to bracket the target sequence. Briefly, two reactions were
run for each of the two possible products. The DNA was denatured at 94C
for two minutes. Thirty cycles of primer annealing, polymerization by taq
polymerase, and denaturation were performed for thirty seconds each at
57C, 72C and 94C, respectively. PCR products were radioactively
labeled by incorporation of dCTP containing 32P and visualized by
polyacrylamide gel electrophoresis and autoradiography. Band intensities
for both mutant and wild-type reactions were quantified by Kodak Digital
Science 1D Image Analysis software. Mutant band intensity was
expressed as a percentage of wild-type band intensity for each sample.
9


2.7 Data correlation
In order to compare trends in biomarkers that differ in their units of
measurement, a common, comparative unit was adopted. At the end of the
data acquisition period, all analytical data were expressed in terms of
percent change of the experimental group from baseline conditions as
determined by the negative control group.
2.8 Reagents
All reagents used were of highest available purity and are available
from Sigma (St. Louis, MO) unless otherwise specified below. Grape seed
extract was acquired from a local commercial source. Trypsin and FBS
were obtained from Life Technologies (Baltimore, MD). Primers for PCR
were obtained from Genosys (Houston, TX).
10


3. Results
The initial goal of this study was to evaluate the ability of a
commercial source grape seed extract (GS) to modulate selected
intermediate biomarkers. Studies of commercial source grape seed extract
have shown it to be a potent antioxidant, and thus, a strong anti-initiator
(unpublished data). In a vigorous, long-term promotion regimen, a grape
seed polyphenol extract has been shown to reduce tumor incidence and
multiplicity in SENCAR mice (Zhao, 1999). This study sought to evaluate
GS during minimal initiation and promotion (five weeks of treatment). The
methods described above were selected so that initiation and promotion
could be evaluated individually, and post-treatment activities of the selected
biomarkers could be observed. Group 3 was used to evaluate the direct
protective effects of GS against initiation events such as oxidative DNA
damage and elevation of other markers indicative of carcinogenesis. Group
4 was used to evaluate any possible anti-promotion effects and further
define the chemopreventive properties of GS. Treatment with GS prior to
treatment with the promoter, but not before treatment with initiator, allowed
for the unique evaluation GS as an anti-promotion agent.
3.1 Inflammatory and hyperplastic responses
Treatment with a strong promoter such as TPA, at the proper dosage
and repetition, will lead to hyperplasia and inflammation in the SENCAR
mouse skin model. These physiological responses are indicative of the
promotion stage of carcinogenesis. Although these responses alone are not
necessarily indicators of cancer development, when coupled with an
li


initiation step these responses become reliable intermediate biomarkers.
Epidermal hyperplasia indicates the ability of the promoter to expand
potentially initiated cells. Dermal hyperplasia and myeloperoxidase activity
are indicative of infiltration of inflammatory cells into the skin.
3.1.1 Epidermal and dermal hyperplasia
The treatments described above were expected to induce a sustained
hyperplastic response in SENCAR mice. Eventually, most of these animals
would develop papillomas and basal and/or squamous cell carcinomas
(DiGiovanni, 1992). Sustained epidermal hyperplasia is an excellent
indicator of the carcinogenic potency of a particular chemical. Conversely, a
lack of epidermal hyperplasia in the skin of an animal treated with a
potentially chemopreventive agent, such as GS, is a good indicator of its
potency.
Figure 3.1 shows the results of the quantification of epidermal and
dermal hyperplasia data. In this minimal, two-stage treatment regimen a
significant increase in both epidermal and dermal hyperplasia was observed
in the DMBA / TPA treated group. However, topical application of GS no
less than ten minutes prior to treatment with DMBA and/or TPA did not
significantly decrease epidermal or dermal hyperplasia. By the ten week
time point (six weeks post-treatment), the thicknesses of both epidermis and
dermis were similar to those in the negative control, acetone treated, group.
12


A
100.00
50.00
0.00
0 weeks 2 weeks 4 weeks 6 weeks 8 weeks 10 weeks
Fig. 3.1. Epidermal and dermal hyperplasia results. (A) Representative pictures of
normal (Acetone) and hyperplastic (DMBA / TPA) skin preparations. (B) Graphic
representation of epidermal and dermal thickness data.* Data show statistically
significant increases over the acetone treated group (students t-test; p-value < 0.05)
13


3.1.2 Myeloperoxidase activity
Inflammatory response is a good intermediate carcinogenesis marker
in the SENCAR mouse model. As previously stated, inflammation and
hyperplasia are the first, observable indicators of skin carcinogenesis. Thus,
a reliable indicator of cutaneous inflammation may also be indicative of
intermediate events in carcinogenesis. MPO is a good indicator of
inflammatory activity because when present in the skin it is only found in
polymorphonuclear leukocytes (PMNs). Tissue macrophages do not contain
MPO. Thus, increased MPO activity indicates the infiltration of inflammatory
cells into the skin. The observation of MPO activity is a quick and easy way
of quantifying inflammatory response in the skin (Trush, 1994).
Inflammation coupled with epidermal hyperplasia are the first observable
indications of carcinogenesis.
The application of GS prior to administration of DMBA and/or TPA did
not decrease MPO activity in the two-stage treatment model employed here
(Figure 3.2). There was a statistically significant increase in MPO activity
observed in DMBA / TPA treated group at the four week time point,
however, there was no significant difference between the GS treated groups
and the positive control group. All experimental groups, and the positive
control group, showed baseline MPO activities within two weeks of
cessation of treatment.
14


Fig. 3.2. MPO activity in epidermal tissue from SENCAR
mice treated with acetone or different combinations of
DM BA, TPA and GS.
* Data show statistically significant increases over the
acetone treated group (students t-test; p-value <0.05).
3.2 DNA Analysis
As discussed
previously, the first step in the
process of carcinogenesis is
an initial insult to the genetic
code, initiation. While
hyperplastic and inflammatory
responses are indicators of
promotion, DNA damage
markers indicate initiation.
While the gross quantification of a particular type of DNA damage is
important, most DNA damage is repaired or results in cell death. Only small
lesions in particular genes allow the cells to live and proliferate and may be
carcinogenic. Both gross oxidative damage and a specific gene mutation
were used as intermediate biomarkers in these studies.
3.2.1 8-OH-dG
The quantification and evaluation of molecular markers indicative of
oxidative damage has gained increasing attention recently. DMBA is
metabolized in the skin to diol epoxides which readily react with DNA and
ultimately leave behind an oxidative lesion on the DNA. Although numerous
oxidative lesions may result from DMBA treatment, elevated 8-OH-dG has
been linked to the development of cancer (Floyd, 1990). Oxidized guanine
in DNA can result in strand breaks or GC -> TA transversions (Nomoto,
1999). The importance of 8-OH-dG as a molecular marker for oxidative
15


Acetone
10 nmol DMBA / 2 ug TPA
2 mg GS *10 nmol DMBA/2 ug
TPA
10 nmol DMBA / 2 mg GS 2 ug
TPA_______________________.
DNA damage is recognized for two
reasons. First, its biological effects
are relatively well defined.
Secondly, techniques for detection
of 8-OH-dG at very low
concentrations (20 fmol) have been
developed.
The results of the
quantification of 8-OH-dG by HPLC
with electrochemical detection are
shown in Figure 3.3. The first
significant decreases in 8-OH-dG
were observed in Group 4 beginning two weeks post-treatment and in Group
3 four weeks post-treatment. These decreases in 8-OH-dG content are not
indicative of any protective effect of GS since they did not occur until the
treatments were completed. Post-treatment decreases in 8-OH-dG may
indicate a residual stimulation of the DNA repair mechanism. However,
further studies would be necessary in order substantiate these claims.
0 weeks 2 weeks A weeks 6 weeks 6 weeks 10 weeks
Fig. 3.3. 8-OH-dG /105 dG ratios for epidermal
tissue from SENCAR mice treated with acetone or
different combinations of DMBA, TPA and GS.
* Data show statistically significant increases over
the acetone treated group. Data show statistically
significant decreases from DMBA / TPA treated
group (students t-test; p-value <0.05).
3.2.2 Ha-ras
The effects of DMBA metabolites on the ras gene are well
documented. The diol epoxide metabolites of DMBA will react with DNA
very specifically causing an A-> T transversion at the second position of
codon 61 (Finch, 1996). This point mutation at codon 61 transforms the ras
proto-oncogene into the Ha-ras oncogene. Because the effects of DMBA
metabolites on this particular gene are so well defined, the Ha-ras oncogene
is an excellent intermediate marker for carcinogenesis in DMBA treated skin.
16


Figure 3.4 shows the results from of the Ha-ras PCR assay. The
initial and two week time points are not shown as no appreciable Ha-ras
mutations were observed. There was a dramatic increase in mutant Ha-ras
band intensity in the DMBA / TPA treated group by the four week time point.
Significant decreases in Ha-ras mutations were detected in the GS treated
A
.1X0 bp-
Ww0
, 1 2 3 4 51 2 3 4 5,
v-----^------v------
Acetone DMBA / TPA
1 234 S,,1 234 5
^ v V
Acetone DMBA/TPA
B
Wild-type Reaction
100 >
90 ti

C Ac* too*
10 nmol 0M8A / 2 ug TPA
c: me cs 10IVTV* DMQA
810 nmol QM8A / 2 GS
vqTPA
up TPA
* AU t*t group* *how*0
algnineaAt lner***a ow
*o*ton* oontrol group.
th*
Mutant Reaction
Fig. 3.4. Mutant / wild-type ratios for A->T transversions at codon 61 of the Ha-ras
oncogene in epidermal tissue from SENCAR mice treated with acetone or different
combinations of DMBA. TPA and GS. (A) Representative autoradiogram showing band
intensities following PCR amplification of mutant and wild-type Ha-ras genes. (B)
Graphic depiction of Ha-ras mutation data, t Data show statistically significant
decreases from DMBA / TPA treated group (students t-test; p-value <0.05).
17


groups at this time point, however, the mutant band intensities rose to levels
that were not significantly less than those of the DMBA / TPA treated group
by the six week time point. Although GS does not appear to have had an
inhibitory effect on Ha-ras codon 61 mutations, it may have slowed the
expansion of initiated cells. These results suggest that GS may have some
anti-promotion activities, however, it is a more secondary inhibitory effect
than that expected of a strong natural antioxidant. Once again, further study
is necessary to substantiate these claims.
3.3 Trend comparisons
The second objective of this study was to compare trends in the
subject intermediate biomarkers over an extended period of carcinogenesis.
In order to make the comparison the percent change in the DMBA / TPA
group from baseline (acetone group) was plotted over time for each variable
(Figure 3.5). These results illustrate two important points. First, all of the
variables showed a significant increase over baseline by the four week time
point, twenty-four hours after administration of the final treatment. For MPO
and 8-OH-dG the only statistically significant increase over baseline values
occurred at this time point. Even with this minimal treatment regimen, all of
the subject intermediate biomarkers showed significant increases.
Secondly, by the end of the experiment, the ten week time point, all elevated
values observed in the subject intermediate biomarkers had returned to
levels that were not statistically significant from baseline values. Though
these intermediate biomarkers are highly predictive of cancer development,
decreased levels do not necessarily indicate a lower risk for cancer
development. It is important to monitor these biomarkers during the time of
exposure to carcinogen.
18


Fig. 3.5. Percent change from baseline with time for all
intermediate biomarkers assayed. For each variable the
results from the DMBA / TPA treated group was compared
to the acetone treated group and a percent change was
calculated.
3.4 Papilloma
development
While the main
purpose of this study was
not to monitor the
development of visible
papillomas and
carcinomas, it is important
to mention that papillomas
were observed at sacrifice
in Groups 2, 3 and 4
beginning at the eight week time point. Intermediate biomarkers are defined
by their predictive nature. In order to show that the subject intermediate
biomarkers are predictive of tumor development it is important to note that
tumors were observed.
19


4. Discussion
The SENCAR mouse model of skin carcinogenesis has proven to be
an excellent tool for assessing the potency of carcinogens and
chemopreventive agents (Slaga, 1987; DiGiovanni, 1992). Cancer
development is a multistep process in which genetic insults are accumulated
over time. The SENCAR mouse skin model has been well defined in terms
of the process of carcinoma development. The definition of this model has
allowed for the discovery of physiological responses and chemical indicators
that mark intermediate outcomes in the process of carcinogenesis. These
intermediate outcome markers have come to be known as intermediate
biomarkers and have proven to be very useful in studies of cancer causation
and prevention. This study proposed to observe trends in the behaviors of
subject intermediate biomarkers over a relatively extended period of
carcinogenesis in a two-step initiation and promotion protocol. Chemical
intervention with a commercial source grape seed extract was also studies
to assess the anti-initiation and anti-promotion properties of this natural
source antioxidant.
The subject intermediate biomarker levels showed significant
increases, over baseline levels, when mice were treated with 10 nmol
DMBA, one time, and 2 pg TPA, twice weekly for four weeks. While this
conclusion seem rather elementary it is important to understand that the
treatment procedure used in this study was considered the minimum
necessary to produce carcinomas in this model. Thus, the observation of
statistically significant increases in the subject intermediate biomarkers
indicates that even at low carcinogenic potency increases in the levels of
20


these biomarkers can be observed. Both initiation indicators, 8-OH-dG and
mutant Ha-ras, and the physiological indicators of promotion, epidermal
hyperplasia and MPO activity, were measurably increased by this treatment
procedure. These biomarkers are highly predictive of the processes that are
necessary for carcinoma development and taken together can predict the
outcome of treatment with a potentially carcinogenic substance.
All of the subject intermediate biomarkers showed significant declines
following cessation of treatment, after the four week time point. Thus, it is
important to understand that while many intermediate biomarkers are
defined by a sustained response, the response is only sustained as long as
active exposure to initiator and/or promoter occurs. In theory, the clonal
expansion of one initiated cell is all that is necessary to cause a carcinoma,
however, researchers must rely on gross tissue damage to detect these
biomarkers. Analysis of the subject intermediate biomarkers is optimized at
the closest posible post-treatment time point.
There are many different natural source antioxidants being studied for
their potential chemopreventive effects. This study utilized a commercial
source grape seed extract based on the results observed in similar studies
(Zhao, 1999; unpublished data). Levels of the subject intermediate
biomarkers were not significantly decreased by topical application of GS
shortly prior to DMBA and/or TPA treatment. Topical application of the
selected GS does not modulate these biomarkers early, i.e. during 5 weeks
of DMBA and TPA treatment, in a two-stage SENCAR mouse model of skin
carcinogenesis. It is important to note that previously published studies
utilized a grape seed polyphenol extract that was synthesized in-house, and
observed only final tumor outcome, not intermediate outcomes.
As was mentioned earlier, and can be seen from the results of this
study, further studies are necessary to define the chemopreventive
21


properties of GS. There is some good evidence from recent experiments
that dietary administration a GS has a greater effect on the subject
intermediate biomarkers. It would also be interesting to look for possible
secondary protective effects, i.e., those that are not directly related to the
antioxidant potential of GS. These so-called secondary effects include the
stimulation of DNA repair mechanisms and slowing of the expansion of
initiated cells, both of which were discussed earlier.
22


References
Boutwell RK. (1964) Some biological aspects of skin carcinogenesis. Prog.
Exp. Tumor Res. 4:207-250.
Bradley PP, Priebat DA, Christensen RD, and Rothstein G. (1982)
Measurement of cutaneous inflammation: estimation of neutrophil content
with an enzyme marker. J. Invest. Dermatol. 78: 206-209.
Cooper GM. (1992) Elements of human cancer. Boston, MA: Jones and
Bartlett Publishers, Inc.
Cooper GM. (1995) Oncogenes. 2nd ed. Boston, MA: Jones and Bartlett
Publishers, Inc.
DiGiovanni J, Slaga TJ, and Boutwell RK. (1980) Comparison of the tumor-
initiating activity of 7,12-dimethylbenz[a]anthracene and benzo[a]pyrene in
female SENCAR and CD-1 mice. Carcinogenesis. 1:381-389.
DiGiovanni J. (1992) Mutlistage carcinogenesis in mouse skin. Pharmac.
Ther. 54: 63-128.
Dipple A., Moschel RC, and Bigger CAH. (1984) Polynuclear aromatic
carcinogens, in Searle, CE (ed). Chemical Carcinogens. 2nd ed.
Washington, DC: American Chemical Society. 1: 41-163.
Finch JS, Albino HE, and Bowden TG. (1996) Quantitation of early clonal
expansion of two mutant 61st codon c-Ha-ras alleles in DMBA/TPA treated
mouse skin by nested PCR/RFLP. Carcinogenesis. 17: 2551-2557.
Floyd RA. (1990) Role of oxygen free radicals in carcinogenesis and brain
ischemia. FASEBJ. 4:2587-2597.
Hanausek M, and Walaszek Z. (1995) Intermediate biomarkers in cancer
prevention and their application in clinical oncology. Cancer Bull. 47: 445-
448.
23


McCabe DR, Acworth IN, Maidt ML, and Floyd RA. (1997) A sensitive and
selective method for the determination of tissue 8-hydroxy-2-
deoxyguanosine using HPLC with electrochemical array detection.
Presented at The 88th Annual Meeting of The American Association for
Cancer Research. San Diego, CA.
Morgan NG. (1989) Cell Signalling. New York: The Guilford Press.
Nelson MA, Futscher BW, Kinsella T, Wymer J, and Bowden GT. (1992)
Detection of mutant Ha-ras genes in chemically initiated mouse skin
epidermis before the development of benign tumors. Proc. Natl. Acad. Sci.
89: 6398-6402.
Nomoto M, Yamaguchi R, Kawamura M, Kohno K, and Kasai FI. (1999)
Analysis of 8-hydroxyguanine in rat kidney genomic DNA after
administration of a renal carcinogen, feric nitrilotriacetate. Carcinogenesis.
20: 837-841.
Slaga TJ, OConnell J, Rotstein J, Patskan G, Morris R, Aldaz M, and Conti
C. (1987) Critical genetic determinants and molecular events in multistage
skin carcinogenesis. Symp. Funmad. Cancer Res. 39: 31-34.
Slaga TJ, Klein-Szanto AJP, Boutwell RK, Stevenson DE, Spitzer HL, and
DiMotto B. (1989) Progress in Clinical and Biological Research. Skin
Carcinogenesis. Mechanisms and Fluman Relevance. New York: Alan R.
Liss, Inc.
Trush MA, Egner PA, and KenslerTW. (1994) Myeloperoxidase as a
biomarker of skin irritation and inflammation. Fd. Chem. Tox. 32: 143-147.
Vogel VG, and Levin B. (1995) Chemoprevention of human cancer-an
overview. Cancer Bull. 47:473-479.
Zhao J, Wang J, Chen Y, and Agarwal R. (1999) Anti-tumor-promoting
activity of a polyphenolic fraction isolated from grape seeds in the mouse
skin two-stage initiation-promotion protocol and identification of procyanidin
B5-3-gallate as the most effective antioxidant constituent. Carcinogenesis.
20(9): 1737-1745.
24


Full Text

PAGE 1

COMPARISONS OF INTERMED I ATE BIOMARKERS DURING CARCINOGENESIS AND INTERVENTION WITH GRAPE SEED EXTRACT by Erick Alan Spears B.A., University of Colorado, 1994 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Arts Biology 2000 [ ; L , .rl ;

PAGE 2

2000 by Erick Alan Spears All rights reserved.

PAGE 3

This thesis for the Master of Arts degree by Erick Alan Spears has been approved by

PAGE 4

Spears, Erick Alan (M. A., Biology) Comparisons of Intermediate Biomarkers During Carcinogenesis and Intervention with Grape Seed Extract Thesis directed by Associate Professor Bradley J. Stith ABSTRACT The SENCAR mouse skin model of carcinogenesis provides the necessary framework for studying the chemical and physiological indicators of carcinogenesis Intermediate biomarkers are chemical indicators or physiological responses that mark intermediate outcomes in carcinogenesis These biomarkers are highly predictive of cancer development. The SENCAR mouse model was used to understand trends in intermediate biomarkers over an extended period of carcinogenesis as well as to assess the inhibitory effects of a natural source antioxidant on selected intermediate biomarkers. Experimental groups of thirty mice were treated one time with 10 nmol DMBA followed by twice weekly treatment with 2 llg TPA for four weeks. Two groups were treated with a commercial source grape seed extract ten minutes prior to TPA, and one of these groups was treated with grape seed extract prior to DMBA. All groups were sacrificed five mice at a time every two weeks beginning just prior to the first TPA treatment. All skin samples were analyzed for selected intermediate biomarkers, epidermal and dermal hyperplasia, myeloperoxidase activity, 8-hydroxy-2 -deoxyguanosine, and Ha-ras oncogene. The commercial source grape seed extract did not modulate intermediate biomarker leve l s in this study. The treatment regimen specified for this study was sufficient to induce significant increases in levels of each of the intermediate biomarkers. While significant increases were observed, by the end of the study biomarker levels had returned to baseline levels. This abstract accurately represents the content of the candidate s thesis recommend its publication. IV

PAGE 5

CONTENTS Figures ...................... ...... ............................................................................. vii Chapter 1 Introduction ........................................................................................ 1 2. Methods ........ .................................................................................... 5 2 1 SENCAR mouse skin carcinogenesis ........ ........................................ 5 2 2 Histological evaluation of epidermal and dermal hyperplasia ............. 6 2.3 Myeloperoxidase activity assay .......................................................... 6 2.4 DNA extraction ................................................................................... 7 2.5 8-Hydroxy-2'-deoxyguanosine analysis .............................................. 8 2.6 Ha-ras oncogene analysis .................................................................. 9 2.7 Data correlation ................ ............................................................... 10 2.8 Reagents .......................................................................................... 10 3. Results ............................................................................................. 11 3 1 Inflammatory and hyperplastic responses ........................................ 11 3.1. 1 Epidermal and dermal hyperplasia ................................................... 12 3.1. 2 Myeloperoxidase activity .................................................................. 14 3.2 DNA analysis ................................................................................... 15 3.2.1 8-0H-dG ....................................... ...... ................... ........... ....... ....... 15 v

PAGE 6

3.2 2 Ha-ras ..... ......................................................................................... 16 3.3 Trend comparisons .......................................................................... 18 3.4 Papilloma development .................................................................... 19 4. Discussion ........................................................................................ 20 References ............................................................ ............................ ..... ... 23 Vl

PAGE 7

FIGURES Figure 3.1 Epidermal and dermal hyperplasia results ................................... 13 3.2 MPO activity in epidermal tissue from SENCAR mice treated with acetone or different combinations of DMBA, TPA and GS ........................................................ .... ... 15 3.3 8-0H-dG I 105 dG ratios for epidermal tissue from SENCAR mice treated with acetone or different combinations of DMBA, TPA and GS ................. ... ........................................... 16 3.4 Mutant I wild-type ratios for T transversions at codon 61 of the Ha-ras oncogene in epidermal tissue from SENCAR mice treated with acetone or different combinations of DMBA, TPA and GS ............. ............................ 17 3 .5 Percent change from baseline with time for all intermediate biomarkers assayed .................... ...... .... .......... ...... ... ................. 19 Vll

PAGE 8

1. Introduction Cancer is a disease characterized by the uncontrolled proliferation of otherwise normal cells (Cooper, 1992). There are many different types of cancer corresponding to the many cell types found in complex animal systems Neoplastic transformation is initiated by an insult to the genetic code, a mutation. Mutations to genes that are directly involved in regulation of the cell cycle may ultimately result in uncontrolled cell proliferation. Once the mutated cell divides, the mutation is "locked-in" to the genome of all progeny Thus, cancer cells divide to form more cancer cells, and eventually invade and damage healthy tissues. Carcinomas, cancers that arise from epithelial tissues, are of particular importance in the field of cancer research Approximately 90% of all human cancers are carcinomas (Cooper, 1995). Due to the rapid cell proliferation that occurs in most normal epithelial tissues, carcinomas are also among the most dangerous cancers. The SENCAR mouse skin model of carcinogenesis provides the necessary framework for studying the chemical and physiological indicators of cancer development. This model of skin carcinogenesis is a stepwise process in which genetic damage is accumulated at each step (DiGiovanni, 1992). The first step, involving an initial insult to the DNA at a target gene, is referred to as initiation. The initiating insult alone is not enough to cause tumorigenesis A second step, known as promotion, must occur in order for the skin cells to become cancerous. The promoting agent causes the cells to undergo clonal expansion and form premalignant lesions called

PAGE 9

papillomas. The final step progression is the accumulation of further genetic changes causing papillomas to become malignant. SENCAR mice are a particular strain genetically developed to be highly sensitive to this type of multistage carcinogenesis using a var i ety of chemical carcinogens (Boutwell 1964 ; DiGiovanni, 1980) Treatment with any known carcinogen at the appropriate dosage and repetition, leads to inflammation and regenerative hyperplasia, dysplasia, papillomas, and finally, basal and/or squamous cell carcinomas (Siaga, 1987; DiGiovanni 1992). A major obstacle to the study of cancer pathogenesis is the fact that tumors begin as normal tissue cells Tumors are not foreign diseases such as pathogenic bacteria that differ greatly in structural and genetic make up from their hosts Often cancers are morphologically and physiologically very similar to normal tissue cells until the late stages of disease. As early detection of cancer has been directly correlated to survival and long-term prognosis the search for early signs of carcinogenesis has become a priority in the field of cancer research These early signals of carcinogenesis have been termed biomarkers (tumor markers, carcinogenesis markers). Biomarkers may be chemical indicators such as unusual proteins, or physiological responses, such as local inflammation and hyperplasia (Siaga 1987; DiGiovanni, 1992). These i ndicators are very useful in cancer research, as tools for early detection of human tumors and in the many in vitro and in vivo models used to study carcinogenesis Intermediate biomarkers occur prior to and are highly predictive of tumor development. They are observed in vitro and in vivo as a means of (1) i dentifying individuals with high susceptibility to cancer (2) measuring compliance with interventions and (3) to marking intermediate outcomes and possibly simplifying experiments (Hanausek, 1995 ) 2

PAGE 10

Cancer prevention has come to the forefront of cancer research recently. Chemoprevention has become a very popular subject in cancer prevention studies. Chemoprevention is defined as the use of chemical agents, either natural or synthetic, to reverse, suppress, or prevent the establishment of a malignant growth (Vogel, 1995). These chemical agents may act to inhibit damage to DNA by blocking the action of carcinogens or interfere with the effects of a promoting agent by suppressing the clonal outgrowth of transformed cells (Cooper, 1992). Antioxidants, chemicals that inhibit oxidative damage to biological molecules, have been recognized as important chemopreventive agents. Oxidative damage has been implicated in both tumor initiation and promotion (Floyd, 1990). Well established in vivo models, such as the SENCAR mouse skin model, have linked oxidative damage to tumor development (DiGiovanni, 1992). In further studies, intermediate biomarker levels and ultimate tumor development have been inhibited by treatment with well-known antioxidants (unpublished results). Recently, natural source antioxidants such as phytochemical extracts have shown promise as anti-carcinogenic agents. Of these phytochemical extracts, the polyphenols found in grape seed extracts have been shown to decrease tumor incidence in a two-stage model of carcinogenesis (Zhao, 1999). In order to evaluate the effects of both initiation and promotion on several intermediate biomarkers, a multistage model of carcinogenesis was employed. The initiating agent used was 7, 12-dimethylbenz[a]anthracene (DMBA). DMBA belongs to a class of chemical carcinogens known as polycyclic aromatic hydrocarbons (PAH). In biological systems, PAHs are metabolized to dial epoxides which bind to DNA and leave behind oxidative lesions (Dipple, 1984 ). Therefore, DMBA is an effective initiator in mouse skin models of multistage carcinogenesis. A one time dose of 10 nmol 3

PAGE 11

DMBA (initiating dose), followed by extensive treatment with a promoting agent is sufficient to cause carcinoma development in SENCAR mice (Boutwell, 1964 ). The most widely studied class of promoters in mouse skin models is the phorbol esters. The most potent phorbol ester is 12-0-tetradecanoylphorbol-13-acetate (TPA; DiGiovanni, 1992). TPA mimics the action of diacylglycerol in cells and initiates the protein kinase C (PKC) pathway. Among the numerous effects of PKC activation is the stimulation of cell growth and proliferation (Morgan, 1989). While a single initiating dose of DMBA is not sufficient to cause tumor formation, four weeks of twice weekly treatment with TPA will ultimately result in carcinoma formation (Siaga, 1989). There is also a safety advantage to this two-stage treatment model. A single, low dose treatment with carcinogen is administered, and further treatments employ a compound, TPA, that is intrinsically non carcinogenic. The purpose of this study was two-fold. First, the behaviors of several intermediate biomarkers were monitored over an extended period of carcinogenesis. Secondly, the effect of co-treatment with a commercial source Grape Seed Extract (GS) on the behaviors of these biomarkers was evaluated The use of a two-stage model allows for the observation of the effects of both initiation and promotion on these intermediate biomarkers 4

PAGE 12

2. Methods 2.1 SENCAR mouse skin carcinogenesis Seven-week old, pathogen free, female SENCAR mice were purchased from the National Cancer Institute (NCI, Frederick, MD) Mice were randomized by weight and separated into four groups of 30 m i ce They were then shaved and topically treated with different combinations of carcinogen antioxidant and solvent. Groups 1 and 2 served as negative and positive controls respectively Group 1 was treated with acetone only as this was the solvent for all treatment compounds. Groups 2,3 and 4 were treated with initiator 10 nmol DMBA, and left untreated for one week Following the one week washout period, 2 1-lg TPA was administered twice weekly for four weeks Acetone treatments were administered at the same times as the DMBA and TPA treatments Groups 3 and 4 were treated with GS no less than ten minutes prior to treatment with TPA. Group 3 was also treated with GS prior to DMBA treatment, while group 4 was not. Animals were sacrificed at two-week intervals beginning just prior to the first treatment with TPA and continuing for six weeks following cessation of treatment. At sacrifice the shaved dorsal skin section was removed. A one square centimeter section was removed from the center of the skin preserved in 10% buffered formalin, and embedded for histological preparation Two 11 mm diameter punches were removed directly anterior and posterior to the h i stological section and frozen in liquid nitrogen. The remaining skin was frozen in liquid nitrogen. All frozen sections were stored at -70C until analysis 5

PAGE 13

2.2 Histological evaluation of epidermal and dermal hyperplasia The formalin preserved sections of the dorsal skin were histologically prepared, stained with hematoxalin and eosin, and examined for epidermal and dermal hyperplasia Epidermal and dermal thicknesses were evaluated using an Olympus BH-2 microscope with a 20.4 mm diameter measuring eyepiece and a micrometer disc. For each sample, 20 measurements were made and an average thickness in micrometers (J.lm) was calculated. 2.3 Myeloperoxidase activity assay Quantification of myeloperoxidase (MPO) activity was carried out using the methods described by Bradley eta/. 1982, and Trush eta/., 1994. Briefly, MPO was extracted from 11 mm diameter punches of skin. The skin was minced and homogenized for twenty seconds in potassium phosphate buffer containing 0 5% hexadecyltr i methylammonium bromide (HTABr), pH 6.0. The homogenized solutions were freeze-thawed and sonicated three times. After the final sonication the suspensions were centrifuged at 15,000 g and 4C for twenty minutes The resulting supernatant was assayed for MPO activity at 460 nm on a Packard, SpectraCount 96 well plate reader. The assay was performed in a potassium phosphate buffer containing 0.5% HTABr 0 .167mg/ml o dianisidine dihydrochloride 0.0005% hydrogen peroxide and 10 J.ll of sample supernatant to a final volume of 300 J.ll. MPO activity in milliunits (mU) was calculated by comparison with curves from known concentrations of MPO standard 0 .310.0 mU 6

PAGE 14

2.4 DNA extraction The final two analyses required DNA purification. DNA was purified by repetitive phenol, chloroform and ether extractions followed by ethanol precipitation. The largest section of frozen skin was thawed and the epidermal layer was isolated and removed by trypsinization. The thawed skin was placed hair side up in a petri dish and scalpel incisions were made approximately 0.5 to 1 0 em apart. The skin was completely covered in 0.25% trypsin and incubated at 37C for two hours. The trypsin reaction was stopped by removal of the trypsin solution and replacement with an equivalent amount of fetal bovine serum (FBS). The epidermal layer was lightly scraped off with a sterile scalpel and transferred to a 13 ml, round bottom, centrifuge tube. The scraped epidermal tissue was frozen in liquid nitrogen and stored at -70C until DNA extraction. The scraped epidermal tissue was gradually thawed and suspended in 3 ml of lysis buffer consisting of 5% SDS, 10 mM NaCI, 10 mM Tris, 10 mM EDT A, and 1 00 flg/ml of Proteinase K The suspension was incubated at room temperature overnight. The first cycle of extraction was performed with 1x volumes of phenol, chloroform I isoamyl alcohol (CIA, 24:1) and ether. Tubes were closed and shaken for ten minutes following the addition of phenol and five minutes following the addition of CIA and ether. After shaking, extracts were spun for ten minutes at 10,000 rpm while in phenol and CIA, and 3,000 rpm while in ether. The organic phases from these extractions were discarded. DNA was precipitated twice by the addition of a 1/1 Ox volume of 4M sodium acetate, and a 2-3x volume of cold 95% ethanol. After the second precipitation, the DNA was dissolved in 600 fll of 7

PAGE 15

1 xTris-EDTA (TE) and stored overnight at 4C. Prior to a second cycle of extraction, a 1110x volume of 1 mglml RNase A was added and samples were incubated at 37C for two hours The second cycle of extraction involved the substitution of phenol I chloroform I isoamyl alcohol (25 : 24 : 1) for phenol and proceeded as described above. After extraction, the samples were again precipitated twice with 4M sodium acetate and 100% ethanol and subsequently dissolved in 500 J.ll of TE. Purified DNA was analyzed spectrophotometrically for purity Optical density (OD) readings were taken at 260 nm and 280 nm. An OD 26010D 280 ratio of 1.8 or higher was considered acceptable purity. The concentration of purified DNA was determined from the OD 260 read i ng. The DNA was also analyzed for degradation by 0. 7% agarose gel electrophoresis 2.5 8-Hydroxy-2'-deoxyguanosine analysis 8-Hydroxy-2'-deoxyguanosine (8-0H-dG) in treated skin was quantified using the isocratic high pressure liquid chromatography (HPLC) method described by McCabe eta/., 1997 Approximately 100 J.lg of DNA from each sample was hydrolyzed with Nuclease P1 and Alkaline Phosphatase and injected on column The chromatographic conditions were similar to those described previously with minor changes. The mobile phase was 100 mM sodium acetate in 4% methanol, pH 5 2. All reagents were HPLC grade and milli-Q water was further purified using a Waters Sep Pak solid phase extraction column !socratic analysis took place on a Shimadzu LC 600 pump connected to a Spectrasystem AS3500 autosampler. A YMCbasic S3, 4.6 mm x 150 mm coi:.Jmn was used for separation The oxid i zed guanine adduct, 8-0HdG, was detected using a 8

PAGE 16

CouiArray electrochemical detection system (ESA, Inc., Chelmsford MA). The cell potentials were set at 50/100/150/230/285/325/400/0. The eighth cell was adapted to a connection with a Shimadzu SPD-6A UV detector set at 254 nm Non-oxidized 2'-deoxyguanosine (dG) was quantified by UV detection. Sample adduct concentrations were calculated from standard curves of 8-0H-dG 0.1 -1 5 pmol, and dG 0 .5-15. 0 nmol. All data were analyzed using ESA CouiArray for Windows software and expressed in terms of the ratio 8 0H-dG/1 05 dG. 2.6 Ha-ras oncogene analysis Mutation of the Ha-ras oncogene in the treated samples was assayed using a polymerase chain reaction (PCR) protocol derived from Nelson et a/., 1992. The mutant and wild -type genes were amplified with mutation specific (5 -CATGGCACTATACTCTICTA) and wild-type specific (5' CATGGCACTATACTCTTCTI) 3 -primers. The same 5 -primer (5' CT AAGCCGTGTTGTTITGCAGGAC) was used in both mutant and wild type react i ons to bracket the target sequence. Briefly two reactions were run for each of the two possible products The DNA was denatured at 94C for two minutes Th i rty cycles of primer annealing polymerization by taq polymerase and denaturation were performed for thirty seconds each at 57 C, 72 C and 94 C respectively. PCR products were radioactively labeled by incorporation of dCTP containing 32P and v i sualized by polyacrylamide gel electrophoresis and autoradiography Band intensities for both mutant and wild-type reactions were quantified by Kodak Digital Science T M 1 D Image Analysis software. Mutant band intensity was expressed as a percentage of wild-type band intensity for each sample 9

PAGE 17

2.7 Data correlation In order to compare trends in biomarkers that differ in their units of measurement a common, comparative unit was adopted At the end of the data acquisition period all analytical data were expressed in terms of percent change of the experimental group from baseline conditions as determined by the negative control group. 2.8 Reagents All reagents used were of highest available purity and are available from Sigma (St. Louis MO) unless otherwise specified below. Grape seed extract was acquired from a local commercial source Trypsin and FBS were obtained from Life Technologies (Baltimore, MD). Primers for PCR were obtained from Genosys (Houston, TX) 10

PAGE 18

3. Results The initial goal of this study was to evaluate the ability of a commercial source grape seed extract (GS) to modulate selected intermediate biomarkers. Studies of commercial source grape seed extract have shown it to be a potent antioxidant, and thus, a strong anti-initiator (unpublished data). In a vigorous, long-term promotion regimen, a grape seed polyphenol extract has been shown to reduce tumor incidence and multiplicity in SENCAR mice (Zhao, 1999). This study sought to evaluate GS during minimal initiation and promotion (five weeks of treatment). The methods described above were selected so that initiation and promotion could be evaluated individually, and post-treatment activities of the selected biomarkers could be observed. Group 3 was used to evaluate the direct protective effects of GS against initiation events such as oxidative DNA damage and elevation of other markers indicative of carcinogenesis. Group 4 was used to evaluate any possible anti-promotion effects and further define the chemopreventive properties of GS. Treatment with GS prior to treatment with the promoter, but not before treatment with initiator, allowed for the unique evaluation GS as an anti-promotion agent. 3.1 Inflammatory and hyperplastic responses Treatment with a strong promoter such as TPA, at the proper dosage and repetition, will lead to hyperplasia and inflammation in the SENCAR mouse skin model. These physiological responses are indicative of the promotion stage of carcinogenesis. Although these responses alone are not necessarily indicators of cancer development, when coupled with an 11

PAGE 19

initiation step these responses become reliable intermediate biomarkers Epidermal hyperplasia indicates the ability of the promoter to expand potentially initiated cells. Dermal hyperplasia and myeloperoxidase activity are indicative of infiltration of inflammatory cells into the skin. 3.1.1 Epidermal and dermal hyperplasia The treatments described above were expected to induce a sustained hyperplastic response in SEN CAR mice Eventually, most of these animals would develop papillomas and basal and/or squamous cell carcinomas (DiGiovanni, 1992) Sustained epidermal hyperplasia is an excellent indicator of the carcinogenic potency of a particular chemical. Conversely, a lack of epidermal hyperplasia in the skin of an animal treated with a potentially chemopreventive agent, such as GS is a good indicator of its potency. Figure 3 1 shows the results of the quantification of epidermal and dermal hyperplasia data. In this minimal, two-stage treatment regimen a significant increase in both epidermal and dermal hyperplasia was observed in the DMBA I TPA treated group However topical application of GS no less than ten minutes prior to treatment with DMBA and/or TPA did not significantly decrease epidermal or dermal hyperplasia By the ten week time point (six weeks post-treatment), the thicknesses of both epidermis and dermis were similar to those in the negative control, acetone treated, group 1 2

PAGE 20

A B 100 00 e ::1. 80.00 "' "' ., c: 60.00 >< u :E 1-40 00 ... E Q; 20 00 'ii w 0 00 200 00 '[ -150.00 "' "' c: tl 100 .00 :E 1-... 50 00 E ., 0 0.00 Epidermis Acetone DMBA/TPA * OAcetone 10 nmol DMBA I 2 ug TPA J rl rL _,-%-,-%-I--r-1-r--0 2 mg GS + 10 nmol DMBA 12 ug TPA 0 nmo l DMBA I 2 mg GS + 2 ug TPA -T J_ _,I. 1 "-----'--'---0 weeks 2 weeks 4 weeks 6 weeks 8 weeks 10 weeks Dermis Fig. 3.1. Epidermal and dermal hyperplasia results (A) Representative pictures of normal (Acetone) and hyperp lastic (DMBA I TPA) skin preparations (B) Graphic representation of epidermal and dermal thickness data Data show statistically significant increases over the acetone treated group (student's t-test; p value < 0 05) 13

PAGE 21

3.1.2 Myeloperoxidase activity Inflammatory response is a good intermediate carcinogenesis marker in the SENCAR mouse model. As previously stated, inflammation and hyperplasia are the first, observable indicators of skin carcinogenesis. Thus, a reliable indicator of cutaneous inflammation may also be indicative of intermediate events in carcinogenesis. MPO is a good indicator of inflammatory activity because when present in the skin it is only found in polymorphonuclear leukocytes (PMNs). Tissue macrophages do not contain MPO. Thus, increased MPO activity indicates the infiltration of inflammatory cells into the skin. The observation of MPO activity is a quick and easy way of quantifying inflammatory response in the skin (Trush, 1994 ). Inflammation coupled with epidermal hyperplasia are the first observable indications of carcinogenesis. The application of GS prior to administration of DMBA and/or TPA did not decrease MPO activity in the two-stage treatment model employed here (Figure 3.2). There was a statistically significant increase in MPO activity observed in DMBA I TPA treated group at the four week time point, however, there was no significant difference between the GS treated groups and the positive control group. All experimental groups, and the positive control group, showed baseline MPO activities within two weeks of cessation of treatment. 14

PAGE 22

0 300 ,--------,=========j OAcetone 0 250 -!----!-----t-110 nmol DMBA /2 ug TPA 6_ 02 mg GS + 10 nmol DMBA / 2 ug TPA 0 200 nmol DMBA /2 mg GS + 2 ug TPA ... ... 2. 0 150 +-;-----;---; f--*--+1-----------1 l:' 0 100 ++-+---1 l+ wr'11----+-+----1f----1 < 0 050 =-0 weeks 2 weeks 4 weeks 6 weeks 8 weeks 10 weeks Fig. 3.2 MPO activity in epidermal tissue from SENCAR mice treated with acetone or different combinations of DMBA TPA and GS. Data show statistically significant increases over the acetone treated group (student s !-test; p-value <0 05). 3.2 DNA Analysis As discussed previously, the first step in the process of carcinogenesis is an initial insult to the genetic code, initiation. While hyperplastic and inflammatory responses are indicators of promotion, DNA damage markers indicate initiation. While the gross quantification of a particular type of DNA damage is important, most DNA damage is repaired or results in cell death. Only small lesions in particular genes allow the cells to live and proliferate and may be carcinogenic. Both gross oxidative damage and a specific gene mutation were used as intermediate biomarkers in these studies. 3.2.1 8-0H-dG The quantification and evaluation of molecular markers indicative of oxidative damage has gained increasing attention recently. DMBA is metabolized in the skin to diol epoxides which readily react with DNA and ultimately leave behind an oxidative lesion on the DNA. Although numerous oxidative lesions may result from DMBA treatment, elevated 8-0H-dG has been linked to the development of cancer (Floyd, 1990). Oxidized guanine in DNA can result in strand breaks or GC -7 TA transversions (Nomoto, 1999) The importance of 8-0H-dG as a molecular marker for oxidative 15

PAGE 23

Cl .., 25. 00 20.00 OAcetooe 10 nmol DMBA /2 ug TPA 02 mg GS + 1 0 nmol D MBA/ 2 ug TPA 10 nmol DMBA/2 mg GS + 2 ug TPA "o 15.00 Cl "i' I I :X: 10.00 q "' I I I -rrr l 5 .00 0 .00 0 weeks 2 wee ks 4 week s 6 weeks 8 weeks 10 we ek s Fig. 3.3. 8 -0H-dG I 105 dG rati os for epidermal tissue from SENCAR mice treated with acetone or different comb inati ons of DMBA TPA and GS Data show statistically sign ifi cant increases over the acetone treated group. :toata show stat i stically significant decreases from DMBA I TPA treated group (student's test ; p-value <0 05) DNA damage is recognized for two reasons. First, its biological effects are relatively well defined. Secondly, techniques for detection of 8-0HdG at very low concentrations (20 fmol) have been developed. The results of the quantification of 8-0H-dG by HPLC with electrochemical detection are shown in Figure 3 .3. The first significant decreases in 8-0H-dG were observed in Group 4 beginning two weeks post-treatment and in Group 3 four weeks post-treatment. These decreases in 8-0H-dG content are not indicative of any protective effect of GS since they did not occur until the treatments were completed. Post-treatment decreases in 8-0H-dG may indicate a residual stimulation of the DNA repair mechanism However, further studies would be necessary in order substantiate these claims 3.2.2 Ha-ras The effects of DMBA metabolites on the ras gene are well documented. The diol epoxide metabolites of DMBA will react with DNA very specifically causing an A-7 T transversion at the second position of codon 61 (Finch, 1996) This point mutation at codon 61 transforms the ras proto-oncogene into the Ha-ras oncogene Because the effects of DMBA metabolites on this particular gene are so well defined the Ha-ras oncogene is an excellent intermediate marker for carcinogenesis in DMBA treated skin 1 6

PAGE 24

Figure 3.4 shows the results from of the Ha-ras PCR assay. The initial and two week time points are not shown as no appreciable Ha-ras mutations were obseNed. There was a dramatic increase in mutant Ha-ras band intensity in the DMBA I TPA treated group by the four week time point. Significant decreases in Ha-ras mutations were detected in the GS treated A 8 Acetone DMBA I TPA Acetone DMBA / TPA Wild-type Reaction Mutant Reaction Fig 3.4. Mutant I wild type ratios for A-7 T transversions at codon 61 of the Ha-ras oncogene in epidermal tissue from S E N CAR mice treated with acetone or different combinat i ons of DMBA TPA and GS (A) Representative autoradiogram showing band intensit i es following PCR amplification of mutant and wild-type Ha-ras genes (B) Graphic depiction of Ha-ras mutation data t Data show statistically significant decreases from DMBA I TPA treated group (student's t-test; p value <0.05) 1 7

PAGE 25

groups at this time point however, the mutant band intensities rose to levels that were not significantly less than those of the DMBA I TPA treated group by the six week time point. Although GS does not appear to have had an inhibitory effect on Ha-ras codon 61 mutations, it may have slowed the expansion of initiated cells. These results suggest that GS may have some anti promotion activities, however, it is a more secondary inhibitory effect than that expected of a strong natural antioxidant. Once again, further study is necessary to substantiate these claims. 3.3 Trend comparisons The second objective of this study was to compare trends in the subject intermediate biomarkers over an extended period of carcinogenesis. In order to make the comparison the percent change in the DMBA I TPA group from baseline (acetone group) was plotted over time for each variable (Figure 3.5) These results illustrate two important points. First, all of the variables showed a significant increase over baseline by the four week time point, twenty-four hours after administration of the final treatment. For MPO and 8-0H-dG the only statistically significant increase over baseline values occurred at this time point. Even with this minimal treatment regimen, all of the subject intermediate biomarkers showed significant increases. Secondly, by the end of the experiment, the ten week time point, all elevated values observed in the subject intermediate biomarkers had returned to levels that were not statistically significant from baseline values Though these intermediate biomarkers are highly predictive of cancer development decreased leve l s do not necessarily indicate a lower risk for cancer development. It is important to monitor these biomarkers during the time of exposure to carcinogen. 18

PAGE 26

c. :::l 300.00 e 25o.oo C) Q) c 200.00 Q) I) < E 150.00 ,g Q) 100.00 01 c tel .r: (.) Ep kkmnal O.nna l Hy pe rpl .. t. Myloper o xidaaa I-OH-dG 2 4 6 8 10 weeks weeks weeks weeks weeks weeks Fig 3.5. Percent change from baseline w i th t ime for all i ntermediate b i omarkers assayed For each v ariable the results from the DMBA I TPA treated group was compared to the aceto ne treated group and a percent change was c alculated 3.4 Papilloma development While the main purpose of this study was not to monitor the development of visible papillomas and carcinomas, it is i mportant to mention that papillomas were observed at sacrifice in Groups 2, 3 and 4 beginning at the eight week time point. Intermediate biomarkers are defined by their predictive nature. In order to show that the subject intermediate biomarkers are predictive of tumor development it is important to note that tumors were observed. 1 9

PAGE 27

4. Discussion The SENCAR mouse model of skin carcinogenesis has proven to be an excellent tool for assessing the potency of carcinogens and chemopreventive agents (Siaga, 1987; DiGiovanni, 1992). Cancer development is a multistep process in which genetic insults are accumulated over time The SENCAR mouse skin model has been well defined in terms of the process of carcinoma development. The definition of this model has allowed for the discovery of physiological responses and chemical indicators that mark intermediate outcomes in the process of carcinogenesis These intermediate outcome markers have come to be known as intermediate biomarkers and have proven to be very useful in studies of cancer causation and prevention. This study proposed to observe trends in the behaviors of subject intermediate biomarkers over a relatively extended period of carcinogenesis in a two-step initiation and promotion protocol. Chemical intervention with a commercial source grape seed extract was also studies to assess the anti-initiation and anti-promotion properties of this natural source antioxidant. The subject intermediate biomarker levels showed significant increases, over baseline levels, when mice were treated with 1 0 nmol DMBA, one time, and 2 TPA twice weekly for four weeks. While this conclusion seem rather elementary it is important to understand that the treatment procedure used in this study was considered the minimum necessary to produce carcinomas in this model. Thus, the observation of statistically significant increases in the subject intermediate biomarkers indicates that even at low carcinogenic potency increases in the levels of 20

PAGE 28

these biomarkers can be observed. Both initiation indicators, 8-0H-dG and mutant Ha-ras, and the physiological indicators of promotion, epidermal hyperplasia and MPO activity, were measurably increased by this treatment procedure. These biomarkers are highly predictive of the processes that are necessary for carcinoma development and taken together can predict the outcome of treatment with a potentially carcinogenic substance. All of the subject intermediate biomarkers showed significant declines following cessation of treatment, after the four week time point. Thus, it is important to understand that while many intermediate biomarkers are defined by a sustained response, the response is only sustained as long as active exposure to initiator and/or promoter occurs In theory, the clonal expansion of one initiated cell is all that is necessary to cause a carcinoma, however, researchers must rely on gross tissue damage to detect these biomarkers Analysis of the subject intermediate biomarkers is optimized at the closest posible post-treatment time point. There are many different natural source antioxidants being studied for their potential chemopreventive effects. This study utilized a commercial source grape seed extract based on the results observed in similar studies (Zhao, 1999; unpublished data). Levels of the subject intermediate biomarkers were not significantly decreased by topical application of GS shortly prior to DMBA and/or TPA treatment. Topical application of the selected GS does not modulate these biomarkers early, i.e. during 5 weeks of DMBA and TPA treatment, in a two-stage SENCAR mouse model of skin carcinogenesis. It is important to note that previously published studies utilized a grape seed polyphenol extract that was synthesized in-house, and observed only final tumor outcome, not intermediate outcomes. As was mentioned earlier and can be seen from the results of this study, further studies are necessary to define the chemopreventive 2 1

PAGE 29

properties of GS There is some good evidence from recent experiments that dietary administration a GS has a greater effect on the subject intermediate biomarkers. It would also be interesting to look for possible secondary protective effects, i.e., those that are not directly related to the antioxidant potential of GS. These so-called secondary effects include the stimulation of DNA repair mechanisms and slowing of the expansion of initiated cells, both of which were discussed earlier. 22

PAGE 30

References Boutwell RK. (1964) Some biological aspects of skin carcinogenesis. Prog. Exp. Tumor Res. 4: 207-250. Bradley PP, Priebat DA, Christensen RD, and Rothstein G. (1982) Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J. Invest. Dermatol. 78: 206-209. Cooper GM. (1992) Elements of human cancer. Boston, MA: Jones and Bartlett Publishers, Inc. Cooper GM. (1995) Oncogenes, 2nd ed. Boston, MA: Jones and Bartlett Publishers, Inc. DiGiovanni J, Slaga T J, and Boutwell RK. (1980) Comparison of the tumor initiating activity of 7, 12-dimethylbenz[a]anthracene and benzo[a]pyrene in female SENCAR and CD-1 mice. Carcinogenesis. 1: 381-389. DiGiovanni J. (1992) Mutlistage carcinogenesis in mouse skin. Pharmac. Ther. 54: 63-128. Dipple A., Moschel RC, and Bigger CAH. (1984) Polynuclear aromatic carcinogens, in Searle, CE (ed). Chemical Carcinogens, 2nd ed. Washington, DC: American Chemical Society. 1: 41-163. Finch JS, Albino HE, and Bowden TG. (1996) Quantitation of early clonal expansion of two mutant 6151 codon c-Ha-ras alleles in DMBAITPA treated mouse skin by nested PCRIRFLP. Carcinogenesis. 17: 2551-2557. Floyd RA. (1990) Role of oxygen free radicals in carcinogenesis and brain ischemia. FASEB J. 4: 2587-2597. Hanausek M, and Walaszek Z. (1995) Intermediate biomarkers in cancer prevention and their application in clinical oncology. Cancer Bull. 4 7: 445448. 23

PAGE 31

McCabe DR Acworth IN, Maidt ML, and Floyd RA. (1997) A sensitive and selective method for the determination of tissue 8 hydroxy 2 deoxyguanosine using HPLC with electrochemical array detection. Presented at The 881 h Annual Meeting of The American Association for Cancer Research San Diego, CA. Morgan NG. (1989) Cell Signalling New York : The Guilford Press Nelson MA, Futscher BW, Kinsella T, Wymer J, and Bowden GT. (1992) Detection of mutant Ha-ras genes in chemically initiated mouse skin epidermis before the development of benign tumors Proc Nat/ Acad. Sci 89: 6398-6402. Nomoto M Yamaguchi R Kawamura M Kohno K and Kasai H. (1999) Analysis of 8-hydroxyguanine in rat kidney genomic DNA after administration of a renal carcinogen feric nitrilotriacetate. Carcinogenesis. 20 : 837-841 Slaga T J, O Connell J Rotstein J Patskan G, Morris R Aldaz M, and Conti C. (1987) Critical genetic determinants and molecular events in multistage skin carcinogenesis Symp Funmad Cancer Res 39 : 31-34 Slaga TJ, Klein-S z anto AJP Boutwell RK Stevenson DE Spitzer HL and Di Motto B (1989) Progress in Clinical and Biological Research, Skin Carcinogenesis, Mechanisms and Human Relevance. New York : Alan R. Liss Inc. Trush MA, Egner PA, and Kensler TW. (1994) Myeloperoxidase as a biomarker of skin irritation and inflammation. Fd. Chern. Tox. 32: 143-147 Vogel VG, and Levin B (1995) Chemoprevention of human cancer an overview. Cancer Bull 47 : 473-479 Zhao J, Wang J, Chen Y and Agarwal R. (1999) Anti-tumor-promoting activity of a polyphenolic fraction isolated from grape seeds in the mouse skin two-stage initiation-promotion protocol and identification of procyanidin B5-3 -gallate as the most effective antioxidant constituent. Carcinogenesis. 20(9) : 1737-1745 24